Many large-scale storage systems are configured as highly-available, distributed storage systems. Such storage systems incorporate a high level of redundancy to improve the availability and accessibility of stored data. For example, a clustered storage system can include a network of controller nodes that control a number of storage devices. A large number of nodes can be configured to have access to the same storage devices, and the nodes themselves can also be communicatively coupled to each another for internode communications. This configuration enables load balancing between the nodes and failover capabilities in the event that a node fails.